competition concentration and their relationship an empirical analysis of the banking industry...

7/30/2019 Competition Concentration and Their Relationship an Empirical Analysis of the Banking Industry 166601

1/24

Competition, concentration andtheir relationship: An empirical analysis

of the banking industry

Jacob A. Bikkera,*, Katharina Haaf

b

a De Nederlandsche Bank, Section Banking and Supervisory Strategies, P.O. Box 98,

NL-1000 AB Amsterdam, The Netherlandsb McGill University, Montreal, Canada

Abstract

This article examines competitive conditions and market structure in the banking industry,

and investigates their interrelationship. Competition is measured using the PanzarRosse

model. In order to distinguish competitive behaviour on local, national and international mar-

kets, for each country, three subsamples are taken: small or local banks, medium-sized banks

and large or international banks. For all 23 countries considered, estimations indicate mono-

polistic competition, competition being weaker in local markets and stronger in international

markets. Subsequently, a relationship for the impact of the market structure on competition is

derived and tested empirically, providing support for the conventional view that concentration

impairs competitiveness.

2002 Elsevier Science B.V. All rights reserved.

JEL classification: F36; G2; L1

Keywords: Local and international banks; Competition; Concentration; Market structure; PanzarRossemodel

1. Introduction

European banking markets are undergoing unprecedented changes, caused by the

deregulation of financial services, the establishment of the economic and monetary

union (EMU) and developments in information technology, which may well turn

Journal of Banking & Finance 26 (2002) 21912214

www.elsevier.com/locate/econbase

*Corresponding author. Tel.: +31-20-524-2352; fax: +31-20-524-3669.

E-mail address: [email protected] (J.A. Bikker).

0378-4266/02/$ - see front matter 2002 Elsevier Science B.V. All rights reserved.

PII: S 0 3 7 8 - 4 2 6 6 ( 0 2 ) 0 0 2 0 5 - 4
http://mail%20to:%[email protected]/http://mail%20to:%[email protected]/http://mail%20to:%[email protected]/


2/24

out to be dramatic. Many of these changes will have vast implications for competi-

tion and concentration in the banking and financial sectors. One of the consequences

is already apparent in the recent wave of mergers in the European banking industry.

This process of concentration may affect competition, in particular on local marketsfor retail banking services. Questions may arise such as: Should concentration be

slowed down? or Are additional measures needed to ensure sufficient competition

in local retail markets? Besides, increased concentration and the size of the new

global players may cause concerns about financial stability. In order to judge the im-

plications of these developments, one has to examine the banking industry s current

market structure, to determine the degree of competition, and to investigate the im-

pact the consolidation is likely to have on the market structure and the behaviour of

banks. In recent years, however, relatively few empirical studies have examined com-

petition and concentration in European banking markets. This article seeks to mea-

sure the degree of competition in the European banking markets, and to investigatethe impact of concentration on competition. Moreover, it attempts to compare the

situation in Europe with that in the US and other countries.

The literature on the measurement of competition may be divided into two main-

streams, called the structural and the non-structural approach. 1 The structural ap-

proach to model competition includes the structureconductperformance (SCP)

paradigm and the efficiency hypothesis, as well as a number of formal approaches

with roots in industrial organisation theory. The SCP paradigm investigates whether

a highly concentrated market causes collusive behaviour among larger banks result-

ing in superior market performance; whereas the efficiency hypothesis tests whether

it is the efficiency of larger banks that makes for enhanced performance. In reaction

to the theoretical and empirical deficiencies of the structural models, non-structural

models of competitive behaviour have been developed namely the Iwata model, the

Bresnahan model, and the Panzar and Rosse (PR) model. These New Empirical In-

dustrial Organisation approaches measure competition and emphasise the analysis of

the competitive conduct of banks without using explicit information about the struc-

ture of the market. In this article we will use one of these non-structural models, the

PR model, to assess the degree of competition in a large number of countries. One

of the structural approaches, the SCP paradigm, provides a theoretical relationship

between market structure (concentration) and conduct (competition) which, in theempirical banking literature, is ignored. This article fills in this gap by using the

PR models measure of competition to test this relationship empirically.

Ideally, an evaluation of competitive conditions and the degree of concentration in

the banking industry should begin by rigorously defining the market under consider-

ation. The relevant market consists of all suppliers of a particular banking service, in-

cluding actual or potential competitors, and it has a product dimension and a

geographical dimension. The product definition of a market is based on the equality

of the products as regards their ability to fulfil specific consumer wants. The geo-

graphical boundaries of a market are determined by actual and potential contacts

1 For an overview, see Bikker and Haaf (2001).

2192 J.A. Bikker, K. Haaf / Journal of Banking & Finance 26 (2002) 21912214


3/24

between actual and potential market participants. These boundaries depend on the

products involved: for retail banking, the local dimension of a market is relevant

while the regional or international dimension is relevant for corporate banking.

The desirability to define product and (smaller-scale) geographical markets makesit harder to apply competition and concentration models to the banking industry, es-

pecially given the shortage in (European) data with respect to specific banking prod-

ucts or local regions.

This article attempts to solve this problem to some extent by applying the PR

model to samples of banks of various sizes, under the assumption that small banks

operate mostly at a local scale and that large banks compete more than other banks

at the international level, while medium-sized banks occupy an intermediate posi-

tion. Furthermore, retail banking is assumed to be concentrated mostly in small

banks while corporate banking occurs more at large banks. Banking behaviour in

geographical markets of various sizes is observed indirectly, through the data of in-dividual banks used in the PR model. Of course, this is only a first step in the right

direction, but we do acquire information about the effect of (the size of) geographical

markets on competition.

The plan of this article is as follows. Section 2 introduces and explains the PR

approach. Section 3 applies this model to banks in 23 industrialised countries. For

each country, four samples are taken: small, medium-sized and large banks and,

finally, all banks. This section also displays various concentration indices and applies

them to the (same) 23 industrialised countries. Finally, the relationship between com-

petition and concentration is tested empirically. The ultimate section summarises

and draws conclusions.

2. The Panzar and Rosse approach

Rosse and Panzar (1977) and Panzar and Rosse (1987) formulated simple models

for oligopolistic, competitive and monopolistic markets and developed a test to dis-

criminate between these models. This test is based on properties of a reduced-form

revenue equation at the firm or bank level and uses a test statistic H, which, under

certain assumptions, can serve as a measure of competitive behaviour of banks. The

test is derived from a general banking market model, which determines equilibriumoutput and the equilibrium number of banks, by maximising profits at both the bank

level and the industry level. This implies, first, that bank i maximises its profits,

where marginal revenue equals marginal cost:

R0ixi; n;zi C0ixi;wi; ti 0: 1

Ri refers to revenues and Ci to costs of bank i(the prime denoting marginal), xi is the

output of bank i, n is the number of banks, wi is a vector of m factor input prices of

bank i, zi is a vector of exogenous variables that shift the banks revenue function, tiis a vector of exogenous variables that shift the banks cost function. Secondly, at the

market level, it means that, in equilibrium, the zero profit constraint holds:

Ri x; n;z Ci x

;w; t 0: 2

J.A. Bikker, K. Haaf / Journal of Banking & Finance 26 (2002) 21912214 2193


4/24

Variables marked with an asterisk () represent equilibrium values. Market power ismeasured by the extent to which a change in factor input prices (dwki) is reflected in

the equilibrium revenues (dRi ) earned by bank i. Panzar and Rosse define a measure

of competition H as the sum of the elasticities of the reduced-form revenues withrespect to factor prices: 2

H Xmk1

oRiowki

wkiRi

: 3

The first market model Panzar and Rosse investigated describes monopoly. The

monopoly analysis includes the case of price-taking competitive firms, as long as the

prices they face are truly exogenous, that is, as long as their equilibrium values are

unaffected by changes in the other exogenous variables in the model. An empirical

refutation of monopoly constitutes a rejection of the assumption that the revenuesof the banks in question are independent of the decisions made by their actual or

potential rivals. Panzar and Rosse proved that under monopoly, an increase in input

prices will increase marginal costs, reduce equilibrium output and subsequently re-

duce revenues; hence H will be zero or negative. This is a very generalised result,

requiring little beyond the profit maximisation hypothesis itself. Along similar lines,

Vesala (1995) proves that the same result holds for monopolistic competition

without the threat of entry, i.e. with a fixed number of banks. Thus, this case also

falls under what we call monopoly. In the case where the monopolist faces a de-

mand curve of constant price elasticity e > 1 and where a constant returns to scaleCobbDouglas technology is employed, Panzar and Rosse proved that His equal to

e 1. Hence apart from the sign, the magnitude of H may also be of importance, asH yields an estimate of the Lerner index of monopoly power L e 1=e H=H 1.

Three other commonly employed models for an industrial market investigated by

Panzar and Rosse are monopolistic competition and perfect competition and conjec-

tural variation oligopoly, all of which happen to be consistent with positive values

for H. In these models, the revenue function of individual banks depends upon

the decisions made by its actual or potential rivals. For monopolistic and perfect

competition, the analysis is based on the comparative statics properties of theChamberlinian equilibrium model. This model introduces interdependence into

banks structural revenue equations via the hypothesis that, in equilibrium, free entry

and exit results in zero profits. Under a set of general assumptions, 3 it can be proved

that under monopolistic competition, H6 1. Positive values of H indicate that the

data are consistent with monopolistic competition but not with individual profit

maximisation as under monopoly conditions. In other words, banks produce more

and the price is less than would be optimal in each individual case. A priori, mono-

polistic competition is most plausible for characterising the interaction between

2 See Panzar and Rosse (1987) or Vesala (1995) for details of the formal derivation of H.3 One of the assumptions is that the market is in a long-run equilibrium. This assumption can be tested

empirically.



5/24

banks, as it recognises the existence of product differentiation and is consistent with

the observation that banks tend to differ with respect to product quality variables

and advertising, although their core business is fairly homogeneous.

In the limit case of the monopolistic competition model, where banks products areregarded as perfect substitutes of one another, the Chamberlinian model produces the

perfectly competitive solution, as demand elasticity approaches infinity. In this per-

fect competition case, H 1. An increase in input prices raises both marginal and av-erage costs without under certain conditions altering the optimal output of any

individual firm. Exit of some firms increases the demand faced by each of the remain-

ing firms, leading to an increase in prices and revenues equivalent to the rise in costs.

Finally, analysing the conjectural variation oligopoly case, Panzar and Rosse

show that strategic interactions among a fixed number of banks may also be consis-

tent with positive values of H. In general, the value of H is not restricted. In the

special case of perfect collusion oligopoly or a perfect cartel, the value of H is non-positive, similar to the monopoly model. Table 1 summarises the discriminatory

power of H.

The Chamberlinian equilibrium model described above provides a simple link be-

tween H and the number of banks, so between market behaviour and market struc-

ture. The model is based on free entry of banks and determines not only the output

level but also the equilibrium number of banks. Vesala (1995) proves that H is an

increasing function of the demand elasticity e, that is, the less market power is exer-

cised on the part of banks, the higher H becomes. This implies that H is not used

solely to reject certain types of market behaviour, but that its magnitude serves as

a measure of competition. One of the general assumptions underlying the Chamber-

linian equilibrium model mentioned above is that the elasticity of perceived demand

facing the individual firm, ex; n;w, is a non-decreasing function of the number ofrival banks. Panzar and Rosse call this a standard assumption, eminently plausible

and almost a truism. Vesalas result and this assumption together provide a positive

(theoretical) relationship between H and the number of banks, or in a more loose

interpretation an inverse relationship between H and banking concentration.

2.1. The empirical PR model

The empirical application of the PR approach assumes a log-linear marginal cost

function (dropping subscripts referring to bank i)

Table 1

Discriminatory power of H

Values of H Competitive environment

H60 Monopoly equilibrium: each bank operates independently as under monopoly

profit maximisation conditions (His a decreasing function of the perceived demand

elasticity) or perfect cartel.

0 < H< 1 Monopolistic competition free entry equilibrium (His an increasing function of theperceived demand elasticity).

H 1 Perfect competition. Free entry equilibrium with full efficient capacity utilisation.



6/24

ln MC a0 a1 ln OUT Xmi1

bi ln FIPi Xpj1

cj ln EXCOSTj 4

where OUT is output of the bank, FIP are the factor input prices (regarding e.g.funding, personnel expenses and other non-interest expenses) and EXCOST are other

variables, exogenous to the cost function Ci (t in Eq. (1)). Equally, the underlying

marginal revenue function has been assumed to be log-linear of the form

ln MR d0 d1 ln OUT Xqk1

fk ln EXREVk 5

where EXREV are variables related to the bank-specific demand function (z in Eq.

(1)). For a profit-maximising bank, marginal costs equal marginal revenues in

equilibrium, yielding the equilibrium value for output (denoted by an asterisk):

ln OUT a0

d0

Xmi1

bi ln FIPi Xpj1

cj ln EXCOSTj Xqk1

fk ln EXREVk

!,

d1 a1: 6

The reduced-form equation for revenues of bank i is the product of the equilibrium

values of output of bank i and the common price level, determined by the inverse-

demand equation, which reads, in logarithms, as ln p n g ln P

i OUT

i .In the empirical analysis, the following operationalisation of the reduced-formrevenue equation is used:

ln INTR a b ln AFR c ln PPE d ln PCE X

fj ln BSFj

g ln OI e 7

where INTR is the ratio of total interest revenue to the total balance sheet, 4 AFR is

the ratio of annual interest expenses to total funds, or the average funding rate, PPE

is the ratio of personnel expenses to the total balance sheet, or the (approximated)

price of personnel expenses, PCE is the ratio of physical capital expenditure andother expenses to fixed assets, or the (approximated) price of capital expenditure,

BSF are bank specific exogenous factors (without explicit reference to their origin

from the cost or revenue function), OI is the ratio of other income to the total

balance sheet, and e is a stochastic error term. AFR, PPE and PCE are the unit

prices of the inputs of the banks: funds, labour and capital, or proxies of these prices.

In the notation of Eq. (7), the H statistic is given by b c d. In order to verifywhether the competitive structure has changed over time as a result of liberalisation

and deregulation, model (7) is applied to a pooled cross-section (across banks) and

4 Here we follow the specification of the dependent variable of Molyneux et al. (1994). Other authors

use unscaled revenues. Re-estimation of the equation with unscaled revenues yields similar results,

particularly if one of the bank-specific factors is total assets.



7/24

time-series analysis over the time span 198898. We are assuming that the long-term

equilibrium market structure underlying the PR analysis shifts gradually over time

due to institutional changes (as mentioned in the introduction) not incorporated into

the model equation. Ignoring market dynamics may lead to imprecise parameterestimates and biased H statistics, which could in turn result in wrong inferences

about the competitive nature of the banking industry. Therefore, we multiply the

elasticities of H by a continuous time-curve model expeTIME:

ln INTR a b ln AFR c ln PPE d ln PCEeeTIME

X

fj ln BSFj g ln OI e: 8

Note that e 0 indicates that H is constant over time. Without this assumption of

gradual change, the results may be implausibly erratic, as found by Molyneux et al.(1994), who applied the PR model to a series of subsequent years.

The dependent variable is the ratio of total interest revenue to the total balance

sheet, as in Molyneux et al. (1994). The decision to consider only the interest part

of the total revenue of banks is consistent with the underlying notion inherent in

the PR model, that financial intermediation is the core business of most banks.

However, Shaffer (1982) and Nathan and Neave (1989) took total revenue as their

dependent variable. In our sample, the share of non-interest revenues to total reve-

nues is, on average, only 14%. However, it has increased in recent years, doubling

between 1990 and 1998. In order to account for the influence exerted by the gener-

ation of other income on the models underlying marginal revenue and cost func-tions, we also include the ratio of other income to the total balance sheet (OI) as

an explanatory variable. Actually, the PR model we will apply, Eq. (8), encom-

passes the model of Molyneux et al. (g 0).The ratio of personnel expenses to the number of employees (PENE) could be a

plausible alternative to the ratio of personnel expenses to the total balance sheet(PPE) included in our estimations. However, the former ratio is available for only

a small subset of our sample. Furthermore, empirical exercises reveal that results

based on PENE closely approximate those based on PPE. This is probably due to

the size of the sample used, which makes the results less sensitive to measurement

errors. The ratio of physical capital and other expenses to fixed assets is a proxy

of the price of capital. 5 In particular, the balance sheet item fixed assets appears

to be unrealistically low for some banks. However, the exclusion of outliers or a cor-

rection for fixed assets, such as applied by Resti (1997), did not lead to remarkable

changes in the estimation results.

Bank-specific factors (BSF) are additional explanatory variables which reflect dif-

ferences in risks, costs, size and structures of banks and should, at least theoretically,

stem from the marginal revenue and cost functions underlying the empirical PR Eq.

(8). The risk component can be proxied by the ratio of risk capital or equity to total

assets (EQ), the ratio of loans to total assets (LO) and the ratio of non-performing

5Capital expenses includes the cost of premises, equipment and information technology.



8/24

loans to total loans (NPL). More than one variable for risk is considered, as there are

cases where one of these variables may be unavailable for a particular bank. The

ratio of interbank deposits to total customers and short-term funding (BDEP)

and the ratio of demand deposits from customers to total customer and short-termfunding (DDC) are used to capture differences in the deposit mix. Correspondent

bank activities are taken into consideration when the ratio of cash and due from de-

pository institutions (or banks) to total deposits (CDFB) is included. Total assets

(TA) are used as a scaling factor.

A positive parameter for LO is expected, because more loans reflect more potential

interest rate income. The coefficient for OI is probably negative as the generation of

other income may be at the expense of interest income. Regarding the signs of the

coefficients of the other explanatory variables, several writers hold conflicting theo-

ries 6 while others do not have a priori expectations.

3. Empirical results

3.1. Competition in the banking industry

The PR model has been applied to banks from 23 European and non-European

countries, as listed in Table 2. The data have been obtained from the database of the

International Bank Credit Analysis Ltd (Fitch-IBCA), a London-based bank credit

rating agency. In principle, data from individual banks are used for the years 1988

98, but the actual starting dates of the samples vary across countries. 7 For each

country, Table 2 reports the number of banks and available number of observa-

tions. 8 The total number of banks is 5444 and the total number of observations is

almost 29,000. Hence, on average, the sample includes more than 5 observations

(in fact years) for each bank, since some of the observations are lacking due to

non-reporting of (all relevant) data by banks in their annual report, mergers or

new entries in the sample period.

For each country, the model has been adopted to a sample of all banks, as well as

to subsamples of small banks, medium-sized banks and large banks, respectively.

This partition into small, medium-sized and large is based on total assets of thebanks: for each year, the smallest 50% of all banks of the world-wide sample consti-

tute the small-banks sample, the largest 10% of all banks constitute the large-bank

sample, whereas the remainder make up the medium-sized sample. The large-bank

6 For example, Molyneux et al. (1994) expect a negative coefficient for EQ, because less equity implies

more leverage and hence more interest income. However, on the other hand, capital requirements increase

proportionally with the risk on loans and investment portfolios, suggesting a positive coefficient.7 Data of earlier years would be less useful due to serious free entry restrictions in European countries.8 Note that ignoring observations of non-financial institutions, which also provide financial interme-

diation in some subdivision of the banking market, does not distort the current analysis, as the actual

(overall) competitive conditions are observed directly, irrespective of the providers of intermediation

services.



9/24

sample was kept relatively small to ensure that only the really large banks are in-

cluded. The final numbers of observations are affected by the availability of data,

which actually appears to be correlated with the size of the bank. Of course, the size

distribution differs across the countries, see Table 2. An alternative would be to make

the size subdivision per country. However, the border between small and medium-

sized banks in, say, 1997 would then range from US$ 166 million in Denmark toUS$ 12,660 million in South Korea and between medium-sized and large banks

would range from US$ 2797 million in Germany to US$ 111,280 million in Canada.

This alternative is less desirable because it makes it very hard to compare subsamples

across countries 9 and, therefore, has not been considered.

Due to the small number of assumptions (mainly the cost minimisation and dual-

ity hypotheses) underlying the PR approach, the validity of the test is fairly general.

Nevertheless, a few caveats are in order. As we employ consolidated banking data,

the banking market of country X is defined as the hypothetical market where banks

Table 2

Sample period and number of observations per country

Country Sample

period

No. of

years

No. of

banks

No of observation per bank type

All Small Medium Large

Australia 199198 8 39 185 13 115 57

Austria 198998 10 95 434 226 176 32

Belgium 198998 10 85 479 217 194 68

Canada 198898 11 60 363 158 140 65

Denmark 199098 9 96 578 466 79 33

Finland 199098 9 14 77 10 32 35

France 198898 11 393 2489 812 1334 343

Germany 198898 11 2219 10,987 6765 3764 458

Greece 199098 9 22 102 46 37 19

Ireland 199298 7 35 143 15 112 16

Italy 198898 11 365 1943 813 897 233Japan 198998 10 148 1081 17 432 632

Korea (South) 199298 7 21 63 1 34 28

Luxembourg 199098 9 128 825 333 395 97

Netherlands 199198 8 57 307 99 145 63

New Zealand 199098 9 10 52 9 23 20

Norway 198998 10 39 220 74 120 26

Portugal 199198 8 41 268 70 144 54

Spain 199098 9 154 831 204 458 169

Sweden 198998 10 26 145 18 52 75

Switzerland 198898 11 385 1976 1414 485 77

UK 198998 10 213 1220 518 491 211

US 199198 8 799 4190 1383 2326 481

Total 5444 28,958 13,681 11,985 3292

In % 100.0 47.2 41.4 11.4

9 Then, for instance, (so-called) large banks in Denmark, Germany or Switzerland could be smaller

than (so-called) small banks in Finland, Japan or Korea.



10/24

from country Xare active and not, say, the banking market within the national bor-

ders of that country. Moreover, banks operate in various segments of the market,

both geographically and in terms of banking products and, for that matter, also in

various input markets as well. This remark is particularly true of large universalbanks with sizeable foreign activities. These internationally active banks are obvi-

ously confronted by other competitive forces than small regional banks. The PR re-

sult H reflects only some kind of average over all these market segments. Finally, in

certain segments of the markets, banks face competition from non-bank financial in-

stitutions. However, the PR approach does not require observations of non-banks,

as the H statistic is a direct measure of the degree of competition taking competitive

effects from other institutions in its stride.

A critical feature of the Hstatistic is that the PR approach must be based on ob-

servations that are in long-run equilibrium. An equilibrium test exploits the fact that

in competitive capital markets, risk-adjusted rates of return will be equalised acrossbanks. In such a case, the return rates will not be correlated with input prices. 10 We

find that the hypothesis of equilibrium (H 0) cannot be rejected on the 95% signif-icance level, which justifies the applied methodology.

As an illustration, Appendix A presents tables with the estimation results of the

various bank-size categories for three countries. Tables for the other countries con-

sidered can be found in Bikker and Haaf (2000). 11 For New Zealand and South

Korea, the number of small banks is too small to make adequate estimations. Our

basic approach was to create a model for each country and bank size combination,

which included all selected bank-specific factors. Actually, for some countries, data

are unavailable for part of these variables, or available only for a limited number of

banks. In the latter case, we accepted only a slight reduction in the sample and other-

wise disregarded that particular variable. 12 Finally, BSF were deleted, if their coef-

ficients were not significant. This was done mainly to prevent the number of

observations from being reduced by the extra explanatory variables, and also for

economys sake. For the latter reason, insignificant coefficients of the time-trend

variable were also deleted. Sensitivity analyses confirm that the Hestimates are only

slightly, if at all, affected by the deletion of the non-significant variables.

The crucial variable H is equal to (b c d expeTIME) and, hence, depends on

TIME, provided e 6 0. In the latter cases, H has been calculated for 1991 as well as1997. The coefficient of the average funding rate, b, appears to be most significantand almost invariably positive and, hence, the main contributor to H. The coefficient

representing labour cost, c, is also significant and positive in most cases, but usually

smaller than b. The coefficient of the price of capital expenses, d, varies in size, sign

10 An equilibrium test is provided by Eq. (7), after replacement of the dependent variable by the rate of

return on total assets (ROA) or equity (ROE). H 0 would then indicate equilibrium, whereas H< 0would point to disequilibrium.

11 These tables are available upon request from the authors.12 For this reason, the number of observations of small, medium-sized and large banks of a country do

not necessarily add up to the number of all banks. This would only be the case if the model specification

were the same for all bank-size types.



11/24

and level of significance, and is the least important component of H. The elasticity d

may also be small due to the poorer quality of capital expenses and fixed assets data,

which constitute the price level of capital expenses.

Finally, the coefficient of TIME, e, also varies in size, sign and level of significance.In fact, e is zero (because not significant) in 53% of all cases, indicating no significant

change in the competitive conditions. Where e is non-zero, e is positive in 34 out of

the 43 cases, which indicates that competition increases over time in 80% of these

cases (see Table 3, where His shown for both 1991 and 1997 ife 6 0). In the all-banksample, competition in 1997 was higher than in 1991 in all the non-zero cases, ex-

cept for Japan. However, these changes over time are fairly limited, on average 2.3

basis points (bps), which is much less than expected. Increase in competition is more

often observed for medium-sized and large banks than for small banks or the all-

bank sample, but with lower increases (1.6 bps versus, respectively, 2.3 and 3.3

bps). Growth in competition in EU countries has been higher than in non-EU coun-tries for all-bank sample and small banks, but lower for large banks. The latter result

is remarkable as stronger increases in competition for the EU were expected for all

bank size samples. Presumably, the rise in European competition, which had been

expected for 1997, has not yet materialised in that year.

Loans appear to be the most important BSF, both in terms of occurrence and level

of significance. Apparently, the ratio between loans and total assets, as a proxy of

risk, is an important factor in the total interest-to-income ratio. The signs of loans

and other income are in line with expectation (respectively, positive and negative)

for all country and sample-size combinations, apart from Greece where implausible

signs are found for small and medium-sized banks. In general, the regression results

are highly satisfactory, due in part to the large size of the samples: the estimation of

Happears to be very robust. Its value is hardly affected by specification choices, such

as those regarding the BSFs. Furthermore, the goodness of fit of the regression equa-

tions is satisfactory.

The tables in Appendix A and in Bikker and Haaf (2000) also present the esti-

mated values for H and test results for the hypothesis H 0 and 1. Table 3 reportsthese values for Hfor various bank-size samples and where applicable (e 6 0) forvarious years. The superscripts refer to the test results in the footnotes of the tables

in Appendix A and Bikker and Haaf (2000). Values of H for which the hypothesisH 0 is not rejected at a confidence level of 95% are in italics. Values of Hfor whichthe hypothesis H 1 is not rejected at the 95% (or 99%) level of confidence are inboldface (or boldface italics, respectively). 13

For all-banks samples of all 23 countries, both H 0 (perfect cartel 14) and H 1(perfect competition) are rejected convincingly, i.e. at the 99% level of confidence,

13 Where the probability of the null hypothesis is 5% or more the null hypothesis is accepted or not

rejected; if the probability of the null hypothesis is below 1%, the null hypothesis is rejected, and if the

probability of the null hypothesis is between 1% and 5% we posit that the null hypothesis is rejected at the

(stringent) 99% confidence level.14 In all countries, the number of banks is far greater than 1. Hence, H 0 reflects perfect collusion or

cartel rather than monopoly.



12/24

and for nearly all countries even at the 99.9% level. This would imply monopolistic

competition in all countries, without exception. However, this uniform picture be-

comes more diversified when the banking market is split into segments: the market

for (i) small banks operating mostly on a local scale, (ii) middle-sized banks operat-

ing both locally and nationally, and (iii) large banks which also operate internation-

ally. For small banks in one country, Australia, the hypothesis H 0 cannot berejected, which suggests that this market is characterised by perfect collusion. Note,

however, that this result is based on an unusually small sample size of 13 observa-tions. In any case, this result suggests a lower level of competition. For a number

of bank-size and country combinations, the hypothesis H 1 cannot be rejected,

Table 3

Empirical results for H for various bank-size samples and various years

All banks Small banks Medium-sized

banks

Large banks

1991 1997 1991 1997 1991 1997 1991 1997

Australia 0.50a 0.57a 0.14b 0.67a 0.70a 0.63a 0.68a

Austria 0.87a 0.93b 0.91a 0.89a 0.91c

Belgium 0.89a 0.95b 0.88c 0.86a 0.88a

Canada 0.60a 0.62a 0.74a 0.63a 0.56a 0.60a

Denmark 0.32a 0.36a 0.31a 0.34a 0.75a 1.16b

Finland 0.78a 0.67b 0.76c 0.70a

France 0.70a 0.54a 0.59a 0.74a 0.79a 0.89b

Germany 0.60a 0.63a 0.56a 0.59a 0.68a 0.70a 1.05b 1.03b

Greece 0.76a 0.41b 0.66a 1.01c 0.94c

Ireland 0.65a

0.99b

0.63a

0.93c

Italy 0.82a 0.75a 0.89a 0.86a 0.83a 0.81a

Japan 0.58a 0.54a 0.43b 0.07c 0.11c 0.64a 0.61a

Korea (South) 0.68a 0.72b 0.77c

Luxembourg 0.93a 0.94b 0.94b 0.95b 0.90a 0.91a

Netherlands 0.75a 0.74b 0.87a 0.91c 0.95c

New Zealand 0.86a 1.11b 1.13c 0.86d

Norway 0.74a 0.77a 0.80a 0.71a 0.75a 0.66a 0.71a

Portugal 0.83a 0.84b 0.88a 0.84a 0.91c

Spain 0.55a 0.62a 0.56a 0.64a 0.52a 0.59a 0.61a 0.66a

Sweden 0.80a 0.84b 0.69a 0.76a 0.95c

Switzerland 0.55a 0.58a 0.51a 0.54a 0.95b 0.92c 1.01d

United Kingdom 0.61a 0.64a 0.41a 0.81a 0.85a 1.20aUnited States 0.54a 0.56a 0.61a 0.62a 0.53a 0.54a 0.68a 0.72a

Averages 0.70 0.64 0.75 0.86

Maximum 0.93 0.99 1.12 1.20

Minimum 0.34 0.14 0.09 0.58Avgs Europe 0.72 0.68 0.79 0.91

Avgs RoW 0.63 0.41 0.63 0.70

For each country, the superscripts refer to the acceptance or rejection of the null hypothesis H 0 andH 1, as is explained in the footnotes of the respective tables in Appendix A and Bikker and Haaf (2000).Averages: Where the underlying model includes a time trend, averages are taken over H-values of 1991 and

1997. Subsequently, averages are taken over the 23 countries.



13/24

implying that these markets may be characterised by perfect competition. This holds

in particular for a number of the large-bank markets. For Danish large banks, H is

significant larger than 1, which may indicate a deviating market structure, such as

conjectural variation oligopoly (see Section 2).A shortcoming of the PR test may be the fact that it is single-tailed, in the sense

that a positive value rejects any form of imperfect competition (under the assump-

tion of profit-maximising behaviour and at least short-term competition), whereas

a negative value is consistent with various types of market power. 15 However, this

shortcoming is of little consequence for this article, as H6 0 is rejected in all cases

but one.

Based on the result of Vesala (1995, p. 56), we interpret H between 0 and 1 as a

continuous measure of the level of competition, in the sense that higher values of

H indicate stronger competition than lower values. The values of H are comparable

across national or bank-size markets if the revenue equation, the perceived demandelasticity and its sensitivity to the number of banks (oe=on) are identical for the var-ious markets. 16 In particular, the latter cannot be observed, so the following conclu-

sions are under reservations that the necessary assumptions hold. The averages

across all countries (bottom rows of Table 3) make clear that H is substantially be-

low average for small-bank markets (0.64), somewhat greater for medium-sized bank

markets (0.75) and greatest for large-bank markets (0.86). Apparently, in line with

expectations, smaller banks operate in a less competitive environment than larger

banks, or, put differently, local markets are less competitive than national and inter-

national markets. The 0.640.86 spread should be seen as the lower limit of the actual

difference in competition between local and international markets, as small banks do

not solely operate on local markets but may also be active on the national markets;

similarly large banks do not exclusively operate on international markets. The differ-

ence in the degree of competition between local and international markets is not only

reflected in the overall averages, but also in many of the national figures. The values

ofHfor small-bank markets range from 0.14 to 0.99, whereas for large banks theyrange from 0.58 to above 1.

In Europe, all large banks appear to operate in a highly competitive environment.

Exceptions are two Scandinavian countries (Finland and Norway) and Spain with H

values around 0.7. Competition among smaller banks is weak in Greece, Denmarkand the UK, and limited in France, Germany, Spain and Switzerland. In general,

competition seems to be weaker in non-European countries. In the US, Canada

and Australia, for instance, H ranges from 0.5 to 0.7, and in Europe from 0.7 to

0.9. Of course, this conclusion does not necessarily hold for all market segments.

In Japan competition is, in fact slightly weaker while in New Zealand and South

Korea it is somewhat stronger. It should be kept in mind, however, that these com-

parisons ofHacross countries are based on more far-reaching assumptions than are

made in the standard PR test (see above). The structure of the European banking

15 See e.g. Bresnahan (1989), Shaffer and DiSalvo (1994) and Toolsema (2002).16 These are sufficient but not necessary requirements.



14/24

industry has altered during the 1980s mostly as in response to domestic deregulation

and in anticipation of EU-wide regulatory changes. One of the major consequences

has been increased competition, see Gual and Neven (1993) and Molyneux et al.

(1996). Gardener and Molyneux (1990) explained how mergers in Europe often con-tributed to the forming of groups of local and regional based banks that could com-

pete effectively with the large, dominant banks. In a number of countries outside

Europe, liberalisation and deregulation took place somewhat later, e.g. in the US,

where the 1994 RiegleNeal Act ending a ban on lending and branch operations

across state borders, and where in 1999, the GlassSteagall Act was repealed, which

has prevented firms from combining banking and insurance activities. These trends

may well have contributed to the observed differences in degrees of competition in-

side and outside Europe.

In order to show what effect our choice for interest revenue as dependent variable

in the PR model had on the outcome, Appendix B contrasts our results with thosebased on total revenue as dependent variable for the Netherlands, where the share of

non-interest revenue in total revenue averages 16%. As the financial intermediation

model underlying the PR theory does not apply to the activities generating non-

interest revenues, it comes as no surprise that the fit of the PR model (measured

by R2) is much lower for total revenue (0.60) than for interest revenue (0.90). In

the total-revenue variant, the estimated value of H is substantially lower, which is

plausible because the input price funding costs is generally not relevant for the

non-interest revenue activities. Nevertheless, this value of H still points to the same

conclusion that the market is characterised by monopolistic competition. Inclusion

of other income as an independent variable hardly affects the outcome. Similar re-

sults are obtained for other countries. All in all, this sensitivity analysis confirms that

total revenue is less appropriate as input for the PR model than interest revenue.

A comparison between our results and those in the literature is shown in Table 4,

which summarises the results of other studies applying the PR model. Shaffer

(1982), in his pioneering study on New York banks, observed monopolistic compe-

tition. For Canadian banks, Nathan and Neave (1989) found perfect competition for

1982 and monopolistic competition for 1983, 1984. Lloyd-Williams et al. (1991) and

Molyneux et al. (1996) revealed perfect collusion for Japan. Molyneux et al. (1994)

obtained values for H which, for 198689, are significantly different from both zeroand unity for France, Germany (except for 1987 when monopoly was found), Spain

and the UK, indicating monopolistic competition. For Italy during 198789, the

monopoly hypothesis could not be rejected. The strong shifts in H over the years

are less plausible, let alone those in market structure. Our pooled time series

cross-section approach ensures less volatile results. Unlike Molyneux et al., Cocco-

rese (1998), who also analysed the Italian banking sector, obtained values for H

which were at the least non-negative and even significantly different from zero.

The value of H also differed significantly from unity, except in 1992 and 1994.

For the Finnish banking industry in the years 198592, Vesala (1995) found con-

sistently positive values ofH, which differed significantly from zero and unity in 1989and 1990 only. De Bandt and Davis (2000) investigated banking markets in France,

Germany and Italy for groups of large and small banks. They obtained estimates of



15/24

H which were significantly different from zero and unity for large banks in all three

countries. The Hstatistics estimated for the small-banks sample indicate monopolis-

tic competition in Italy, and monopoly power in France and Germany. The latter

results are in flat contradiction to our findings. For Switzerland, Rime (1999) ob-

served monopolistic competition. Like Rime, Bikker and Groeneveld (2000) applied

the PR method to 15 EU-countries without, however, distinguishing between size

classes. Their results are rather similar to ours, except for the larger countries, where

they used smaller samples of (only) the largest banks. Their H-values for these coun-

tries may therefore be regarded as overestimations.

In some respects, the empirical PR studies present far from uniform outcomes.

Yet except in the case of Japan and, according to some authors, Italy, they all point

to the existence of monopolistic competition in the countries considered.

3.2. Market structure and competition

Given the current wave of mergers in the EU banking market and the expectation

of continued or even accelerating consolidation, concerns have been voiced as to

competitive conditions in the EU banking markets, especially in some market seg-

ments, such as local and retail markets. More precisely, the question emerges

whether market concentration might affect the conduct of banks or the degree of

competition. Theoretically, the existence of a relationship between market structure

and banks behaviour is indicated by, among others, the PR model. Where, in theliterature, the impact of the banking market structure on bank performance has been

examined exhaustively employing the SCP paradigm the relevance of market

Table 4

PR model results in other studies

Authors Period Countries considered Results

Shaffer (1982) 1979 New York Monopolistic competitionNathan and Neave (1989) 198284 Canada 1982: perfect comp.; 1983,

1984: monopolistic

competition

Lloyd-Williams et al. (1991) 198688 Japan Monopoly

Molyneux et al. (1994) 198689 France, Germany, Italy,

Spain and United

Kingdom

Mon.: Italy; mon. comp.:

France, Germany, Spain, UK

Vesala (1995) 198592 Finland Monopolistic competition for

all but two years

Molyneux et al. (1996) 198688 Japan Monopoly

Coccorese (1998) 198896 Italy Monopolistic competition

Rime (1999) 198794 Switzerland Monopolistic competition

Bikker and Groeneveld

(2000)

198996 15 EU countries Monopolistic competition

De Bandt and Davis (2000) 199296 France, Germany and

Italy

Large banks: mon. comp. in

all countries; small banks:

mon. comp. in Italy, monopoly

in France, Germany



16/24

structure for conduct or competitive conditions has been almost entirely ignored. 17

The present section aims at examining this disregarded relationship and seeks to as-

sess a possible impact of the number of banks and the banking market concentration

on competition.As was observed above, the PR approach provides a link between number of

banks and competition. However, as a description of the market structure, the num-

ber of banks is a rather limited concept. For instance, it fully ignores the size distri-

bution of banks (or inequality) in a given market. As concentration indices, weighted

averages of banks market shares, take both the size distribution and the number of

banks into account, they are often used as a simple proxy of the market structure.

Apart from the number of banks itself, we also use two-frequently applied-types

of such indices as a proxy. 18 The first is the so-called k-bank concentration ratio

(CRk) which takes the market shares of the klargest banks in the market and ignores

the remaining banks in that market. This index is based on the idea that the behav-iour of a market is dominated by a small number of large banks. The second index

we use is the Herfindahl index (HI), which takes market shares as weights, and stres-

ses the importance of larger banks by assigning them a greater weight than smaller

banks. It includes each bank separately and differently, and thereby avoids an arbi-

trary cut-off and insensitivity to the share distribution.

Table 5 presents the 1997 HI and CRk, for k 3, 5 and 10, for all 23 countriesanalysed earlier. Total assets have been taken as the measure of bank size. The value

of the k-bank concentration ratios (for various values ofk) always exceeds the value

of the HI, since the latter gives less prominence to the markets shares (the weights

again being market shares) than the former (unit weights). The results for the various

index values are rather similar, displaying a high degree of correlation. The strongest

correlations are found between CR3 and CR5, CR5 and CR10, and, surprisingly, HI

and CR3. In terms of ranking, the correlation between HI and CR3 is, at 98%, by far

the strongest. This demonstrates that the HI is determined mainly by (the squares of

the market shares of) the large banks, which puts into perspective the alleged draw-

back of the CRk indices vis-aa-vis the HI, i.e. that they ignore the influence of smaller

banks.

For countries where the number of banks in the available sample is low, as in Fin-

land, Korea and New Zealand, results are less reliable. High concentration rates arefound in Denmark, Greece, the Netherlands and Switzerland, where the largest three

banks take more than two-thirds of the total market in terms of total assets. In Can-

ada, concentration is high only when at least five banks are taken into consideration.

Switzerland is the most highly concentrated country according to HI, which may

seem remarkable given the large number of banks in that country. Increasing the

sample of Swiss banks to more than three hardly increases the concentration rate.

Concentration appears to be low in France, Germany, Italy, Luxembourg and the

17 See Calem and Carlino (1991) for an example of the empirical approximation of conduct.18 Both indices, CRk and HI, can be derived as proxies for market structure in theoretical SCP

relationships, see Bikker and Haaf (2001).



17/24

US, where the largest three have a combined share of less than one-third. By all mea-

sures, concentration is lowest in the US. Germany, where the number of banks in our

sample is largest, takes the last place but one in concentration.

All types of indices appear to be inversely correlated to the number of banks. This

is owing to a well-known weakness of concentration indices, namely their depen-

dency on the size of a country or banking market. The smaller the country or thenumber of its banks, the larger its measure of concentration. In the empirical anal-

ysis below, we attempt to solve this problem by taking the number of banks into ac-

count explicitly. Table 5 is based on the Fitch-IBCA data set, used for the PR

analysis in Section 3.1. This sample does not include all banks, which for some coun-

tries might distort the concentration index value. However, this effect is limited as the

ignored market segment consists mainly of the smallest banks. This problem, too, is

considered in the empirical analysis below. Another shortcoming of concentration

indices is that non-bank financial institutions are ignored. As competition from

non-banks is mainly related to some segments of the banking market, such as mort-

gage lending, it is difficult to correct for that in the present broadtotal assets

mea-

sure of concentration. Finally, the Fitch-IBCA data set consists of consolidated

figures and does not distinguish between domestic and foreign operations. For that

Table 5

Concentration indices for 23 countries, based on total assets (1997)

Herfindahl index CR3 CR5 CR10 No. of banks

Australia 0.14 0.57 0.77 0.90 31Austria 0.14 0.53 0.64 0.77 78

Belgium 0.12 0.52 0.75 0.87 79

Canada 0.14 0.54 0.82 0.94 44

Denmark 0.17 0.67 0.80 0.91 91

Finland 0.24 0.73 0.91 1.00 12

France 0.05 0.30 0.45 0.64 336

Germany 0.03 0.22 0.31 0.46 1803

Greece 0.20 0.66 0.82 0.94 22

Ireland 0.17 0.65 0.73 0.84 30

Italy 0.04 0.27 0.40 0.54 331

Japan 0.06 0.39 0.49 0.56 140

Korea (South) 0.11 0.45 0.68 0.96 13

Luxembourg 0.03 0.20 0.30 0.49 118

Netherlands 0.23 0.78 0.87 0.93 45

New Zealand 0.18 0.63 0.90 n.a. 8

Norway 0.12 0.56 0.67 0.81 35

Portugal 0.09 0.40 0.57 0.82 40

Spain 0.08 0.45 0.56 0.69 140

Sweden 0.12 0.53 0.73 0.92 21

Switzerland 0.26 0.72 0.77 0.82 325

United Kingdom 0.06 0.34 0.47 0.68 186

United States 0.02 0.15 0.23 0.38 717

Averages/total 0.12 0.49 0.64 0.73 4645

Standard deviations 0.07 0.18 0.20 0.18



18/24

reason, the concentration index values for small countries with large international

banks are overestimated. 19 This is illustrated in Table 6, where market shares for

total assets, loans and non-bank deposits are presented for the ten banks with the

largest balance-sheet total in the sample.

The market shares of even the largest banks in France, Germany, Japan and the

UK are modest, suggesting that, on a national or market-wide scale, these banks

have limited market power. Of course, the situation may be different in particular

market segments or local areas. Large banks in small countries, such as UBS of Swit-

zerland and ABN-Amro of the Netherlands, do command substantial market shares,

although in the case of ABN-Amro, for instance, domestic business is less than half

of total business, while for UBS the ratio is even less than one in four. After correc-

tion for foreign operations (which, by the way, affects both the denominator and the

numerator of the shares), the market shares for ABN-Amro are almost 30% smaller, 20

though still sizeable. To some extent, the indices (such as CR3) do indeed reflect the

higher concentration in the Netherlands and Switzerland. At the same time, ABN-

Amro and UBS may be facing stiff competition on other (foreign) markets.In order to investigate the relationship between competition and market structure

in the banking industry, we related the Hstatistic for all banks, 21 a measure of com-

petition, to the concentration index (CI) and the logarithm of the number of banks in

the markets (log n) as representatives of the market structure. Despite the various

shortcomings, already noted, which concentration index may have as a proxy of

Table 6

Market shares of the ten largest banks in the sample (1997)

Bank Country Market shares National

Total assets Loans Deposits H CR3

1. UBS AG Swi 0.36 0.46 0.45 1.01 0.72

2. Deutsche Bank AG Ger 0.09 0.09 0.12 1.03 0.22

3. HSBC Holdings Plc UK 0.14 0.14 0.15 1.20 0.34

4. Bayerische Hypo- und

VereinsbankaGer 0.09 0.07 0.04 1.03 0.22

5. Creedit Agricole CA Fra 0.14 0.11 0.13 0.89 0.30

6. Dai-Ichi Kangyo Bank Ltd

DKB

Jap 0.17 0.14 0.12 0.61 0.39

7. ABN Amro Holding N.V. Neth 0.34 0.39 0.41 0.95 0.78

8. Socieetee Geeneerale Fra 0.10 0.10 0.11 0.89 0.30

9. Norinchukin Bank Jap 0.08 0.13 0.11 0.61 0.6110. Sakura Bank Limited Jap 0.16 0.13 0.12 0.61 0.61

a Pro forma.

19 A similar distortion does not occur in the PR analysis, where any link to national borders is absent.

There, the level of competition of a country is the average level of competition on the markets where its

banks operate.20 Not in the 50% region lower, due to the denominator effect.21 It would not be sensible to split the concentration index into indices for small, medium-sized and

large banks. Hence the analysis is based on the all-banks sample of H and the concentration indices.



19/24

the market structure, we nevertheless, for want of anything better, examined how it

acts in the empirical relationship in question. The estimated values of H were taken

from the all-bank sample. Concentration indices are one-dimensional measures tak-

ing account of two dimensions, i.e. the number of banks, indicating the density ofthe banking market, and their size distribution, indicating skewness. Some concen-

tration indices can be rewritten as measures of the distribution and the number of

banks. For instance, the Herfindahl index can be rewritten as 22

HI g20 1=n 9

where g20 is the variation coefficient of the bank-size distribution. In our empirical

analysis, we restore this two-dimensionality, describing the market structure by both

the CI and the number of banks. Here we have used logarithms to scale the variable

n. There is another reason why inclusion of the number of banks makes sense. Below

Table 5, we recorded the dependency of concentration indices on the size of a

country or banking market. By including the number of its banks (as proxy of the

banking market size) in the regression equation, we correct for this distortion. At the

same time, the effect on the CI of a limited sample size (as is the case for some

countries) is compensated for. Hence, the estimated regression is given by

H a0 a1CI a2 log n a3dummy Europe: 10

A dummy variable Europe is included because H is substantially higher for Europe

than for non-European banking markets, which may be due to economic and in-stitutional conditions. The upper part of Table 7 presents regression results for Eq.

(10) for four CIs. 23 For all four regressions, the coefficient of the concentration

index shows the expected negative sign, indicating that competition is decreasing

with increasing market concentration. The significance of CR3 is highest (highest t-

value), while the effect of CR10 on competition is strongest (taking into account the

indices standard deviation). The results support the view that the share of the k

largest banks (k being 3, 5 or 10) rather than the entire size distribution of banks in

a market, is the strongest determinant for the competitive conditions in a market.

In principle, a larger number of banks indicates more potential for competition.

For that reason, one would expect to see a coefficient with a positive sign. Yet, wher-

ever the size distribution is heavily skewed and just a few banks dominate the mar-

ket, a large number of banks merely indicates that there is a broad fringe of

powerless dwarfs. The larger the number of banks, the less opportunity each

22 See Bikker and Haaf (2001, Eq. (2.5)).23 Alternatively, the observations for the 23 countries could also be weighted by log n as a measure (or

proxy) of the respective banking-market size. In this way, we could take account of the magnitude of the

banking market in each country, weighting Germany, France and Italy heavier than, say, Greece, Ireland

and Sweden. At the same time, lower weights would be allocated to countries where the sample size is

limited and the variables are therefore somewhat less reliable, as is the case for Finland, South Korea and

New Zealand. Incidentally, the weighted regression results appear to be similar to the unweighted

outcomes of Table 7, while the level of significance is even substantial higher for CI and log(n).



20/24

has to assert its influence and increase competition. The latter effect appears to be

dominant, as the net effect of the number of banks on competition is negative. 24 Ob-

viously, in situations, typically measured by the CRk indices, where a few large banks

occupy a large share of the market, market power is exercised more heavily. The

dummy variable for Europe proves to be significant for all four regression equations,

indicating the different institutional and economic conditions faced by European

banks, which are disregarded by the other explanatory variables.It is no easy task to find out which differences cause the diverging levels of com-

petition between European and non-European banks. We found that if the share of

bank demand deposits in total assets is included in Eq. (10), the Europe dummy be-

comes insignificant and may be dropped, see the lower part of Table 7. This share

reflects an aspect of banks average funding habits: European banks make more ex-

tensive use of the interbank market for funding. We may have hit on an important

variable here, 25 but a convincing theoretical explanation is lacking. One possible ex-

planation is that where banking and financial markets are more developed, (i) the

interbank market is more developed and (ii) competition is more intense. If this is

true, then the share of bank deposits acts as an indicator of financial sophistication.The above provides evidence that conduct such as competitive behaviour may in-

deed be related to characteristics of the market structure such as concentration and

number of banks. A Wald test, which assesses the significance of the coefficients

of the CI and the number of banks simultaneously, confirms this conclusion. 26

Table 7

Relationship between competition and concentration for the all-bank sample

HI CR3 CR5 CR10

Constant 1.01 (7.0) 1.23 (7.1) 1.31 (5.5) 1.54 (4.6)Concentration Index 0.93 (1.9) 0.53 (2.9) 0.47 (2.3) 0.65 (2.5)Log n 0.07 (2.8) 0.09 (3.6) 0.10 (3.2) 0.11 (3.0)Dummy (Europe) 0.15 (2.3) 0.16 (2.8) 0.14 (2.3) 0.17 (2.9)

R2

0.25 0.37 0.30 0.35

Constant 0.90 (7.0) 1.05 (6.6) 1.14 (5.3) 1.33 (4.5)

Concentration Index 0.71 (1.8) 0.38 (2.4) 0.37 (2.1) 0.51 (2.4)Log n 0.07 (3.1) 0.08 (3.6) 0.09 (3.3) 0.10 (3.1)Share of bank deposits 0.85 (3.6) 0.83 (3.7) 0.80 (3.5) 0.89 (4.0)

R2

0.42 0.48 0.45 0.50

Note: T-values in parenthesis. The critical value of the one sided t-value test is 1.73 and that of the two-

sided t-value test is 2.09.

24 An alternative formulation is that the effect of the number of banks, as taken into account by the CI,

is overestimated (implying low values for the CI value in large countries and vice versa).25 As suggested, for instance, by the better fit in Table 7, when the share of interbank deposits is

included.26 For all (eight) equations in Table 7, and also in the weighted regressions (not shown), a hypothetical

value of zero for both coefficients is rejected at the 95% level of confidence.



21/24

The evidence is only modest, however, as Eq. (10) is not very robust. For instance, if

the Europe dummy is deleted, the significance of the other parameters drops dramat-

ically. Nevertheless, the continuing process of consolidation in the banking industry

may raise the policy makers concern about competitive conditions in the bankingmarkets. This concern, however, is very often related not so much to entire markets

as to market segments, such as, for instance, deposits or certain geographical areas.

Unfortunately, the limited availability of data makes it impossible to make more re-

fined analyses.

4. Conclusions

The ongoing dramatic structural changes in the banking industry, particularly in

Europe, may affect competition, especially on local markets and for banks retailservices. This article sought to assess competitive conditions and concentration in

the banking markets of as many as 23 industrialised countries inside and outside

Europe over approximately 10 years. In addition, it investigated the interaction be-

tween competition and concentration. The PanzarRosse approach has been ap-

plied to obtain a measure of competitive conditions in to 23 countries over a

time span of more than 10 years. The resulting H statistic provides strong evidence

that the banking markets in the industrial world are characterised by monopolistic

competition, but perfect competition cannot be ruled out in some cases.

We have attempted to take account of the geographical and even the product di-

mension of banking operations by defining three sub-markets in terms of bank sizes

for each country and have estimated their degree of competition. Competition is

stronger among large banks operating predominantly in international markets

and weaker among small banks operating mainly in local markets while medium-

sized banks take an intermediate position. In some countries, perfect competition has

been found among large banks. For a number of countries, estimates of the Hstatistic

over time indicate a significant increase in competition. Competition seems to be

somewhat stronger in Europe than in countries like the US, Canada and Japan.

Thanks to the large sample and the pooled regression approach, our results are fairly

robust. Generally speaking, our findings are in keeping with comparable studies in theliterature, which also point to monopolistic competition in most countries.

Concentration in the banking markets of 23 industrialised countries was measured

using various k-bank concentration ratios and the Herfindahl index. Empirical stud-

ies concerning the impact of market structure on banks conduct are rare. In order to

investigate this relationship, the estimated H-values indicating competition are used

as proxies of conduct and are related both to the concentration indices considered

and to the absolute number of banks operating in these markets, acting together

as a proxy of the market structure. The impact of both market structure measures

on competition appears to be significant, most markedly so when the k-bank concen-

tration indices are used. The latter confirms the observation that a few large (cartel)banks can restrict competition and that a multitude of fringe competitors is unable to

engender competition.



22/24

Appendix A. Estimation results of the PR models for four countries and various size

classes (Tables 810)

Table 9

Empirical results for the Netherlands (199198)

All banks Small banks Medium-sized banks Large banks

Coefficient t-values Coefficient t-values Coefficient t-values Coefficient t-values

Funding rate 0.75 48.5 0.70 23.1 0.85 48.9 0.67 19.1

Wage rate 0.08 5.0 0.01 0.2 0.06 4.7 0.09 9.2

Capital price 0.07 2.9 0.05 0.3 0.05 2.2 0.12 4.3

Time 6.80 4.4

Loans ratio 0.04 3.4 0.04 4.76 0.27 4.8

Other income 0.05 5.1 0.06 3.1 0.03 2.9

Total assets 0.01 2.3 0.03 3.2 0.02 3.1

Equity 0.12 8.7 0.14 5.6 0.07 4.5

Intercept 0.73 8.8 1.31 7.9 0.44 4.5 0.20 1.6

Adj. R2 0.90 0.90 0.95 0.95

No. of observa-

tions

296 96 139 61

H (199197) 0.75a 0.74b 0.87a 0.91c 0.95c

aH 0 and H 1 rejected (level of confidence 99.9%).bH 1 not rejected (level of confidence 95%); the F-statistic is 2.33 and the probability level (of the null

hypothesis) is 13.0%.cH 1 not rejected (level of confidence 95%); the F-statistic is, respectively, 2.72 and 0.92, and the

probability level (of the null hypothesis) is, respectively, 10.5% and 34.1%.

Table 8Empirical results for Germany (198898)

All banks Small banks Medium-sized banks Large banks


Funding rate 0.45 74.1 0.39 53.7 0.50 50.9 0.92 28.1

Wage rate 0.13 35.9 0.13 27.5 0.16 25.4 0.14 17.2

Capital price 0.00 0.3 0.01 4.8 0.01 2.3 0.00 0.1

Time 7.07 19.0 9.68 18.8 4.72 9.3 2.32 2.6

Loans ratio 0.07 32.0 0.09 33.0 0.04 10.0

Other income 0.05 23.6 0.06 21.0 0.05 13.7

Total assets 0.01 8.9 0.03 4.7

Bank deposits 0.01 6.8 0.00 2.1 0.01 6.0 0.12 12.5Equity 0.04 12.5 0.02 5.3 0.03 5.4 0.22 18.5

Dem. Dep. C. 0.01 4.5 0.02 8.2

Cash & DFB 0.01 7.3 0.01 7.0 0.01 5.4

Intercept 0.64 26.3 0.83 27.6 0.45 11.2 1.05 8.8

Adj. R2 0.69 0.69 0.71 0.84

No. of observa-

tions

10 513 6523 3672 458

H (199197) 0.60a 0.63a 0.56a 0.59a 0.68a 0.70a 1.05b 1.03b

aH 0 and H 1 rejected (level of confidence 99.9%).bH 1 not rejected (level of confidence 95%).



23/24

Appendix B. The dependent variable: Interest revenue versus total revenue (Table 11)

Table 10

Empirical results for the United States (199198)

All banks Small banks Medium-sized

banks

Large banks


Funding rate 0.40 46.7 0.40 29.0 0.40 35.8 0.43 19.6

Wage rate 0.07 13.7 0.18 16.1 0.06 8.4 0.13 8.7

Capital price 0.06 19.7 0.02 3.2 0.06 14.1 0.10 15.4

Time 3.62 4.4 3.08 2.9 3.52 3.0 8.01 4.8

Loans ratio 0.12 20.0 0.09 11.2 0.12 14.1 0.21 18.3

Other income 0.04 6.4

Total assets 0.02 10.1 0.02 4.5

Equity 0.06 7.4 0.03 2.5 0.05 4.7

Non-performing

loans

0.01 4.3 0.01 3.7

Cash & DFB 0.03 11.9 0.02 5.2 0.04 9.6

Intercept 0.76 16.3 0.63 8.3 0.76 11.2 0.02 2.6

Adj. R2 0.51 0.48 0.52 0.70

No. of observa-

tions

3835 1350 2216 463

H (199197) 0.54a 0.56a 0.61a 0.62a 0.53a 0.54a 0.68a 0.72a

aH 0 and H 1 rejected (level of confidence 99.9%).

Table 11

Empirical results for the Netherlands for interest revenue and total revenue (all banks; 199198)

Interest revenue Total revenue

Including other

income

Excluding other

income

Including other

income

Excluding other

income


Funding rate 0.75 48.5 0.76 47.8 0.48 17.6 0.47 17.0

Wage rate 0.08 5.0 0.01 2.0 0.07 2.5 0.16 13.3

Capital price 0.07 2.9 0.01 1.6 0.01 0.7 0.02 1.3Time

Loans ratio 0.04 3.4 0.04 3.4 0.04 1.6 0.02 1.0

Other income 0.05 5.1 0.06 3.5

Total assets 0.01 2.3 0.01 1.4 0.01 1.1 0.00 0.4Equity 0.12 8.7 0.11 7.6 0.07 2.9 0.08 3.3

Intercept 0.73 8.8 0.67 8.0 0.51 3.5 0.50 34

Adj. R2 0.90 0.90 0.59 0.57

No. of observa-

tions

296 296 296 296

H (199197) 0.75a 0.77a 0.56a 0.65a

aH 0 and H 1 rejected (level of confidence 99.9%).



24/24

References

Bikker, J.A., Groeneveld, J.M., 2000. Competition and concentration in the EU banking industry. Kredit

und Kapital 33, 6298.Bikker, J.A., Haaf, K., 2000. Competition, concentration and their relationship: An empirical analysis of

the banking industry. Research Series Supervision no. 30, De Nederlandsche Bank, Amsterdam.

Bikker, J.A., Haaf, K., 2001. Measures of competition and concentration: A review of the literature. De

Nederlandsche Bank, Amsterdam.

Bresnahan, T.F., 1989. Empirical studies of industries with market power. In: Schmalensee, R., Willig,

R.D. (Eds.), Handbook of Industrial Organisation, vol. II. Elsevier, Amsterdam, pp. 10121055.

Calem, P.S., Carlino, G.A., 1991. The concentration/conduct/relationship in bank deposit markets.

Review of Economics and Statistics 73, 268276.

Coccorese, P., 1998. Assessing the competitive conditions in the Italian banking system: Some empirical

evidence. BNL Quarterly Review, no. 205, 171191.

De Bandt, O., Davis, E.P., 2000. Competition, contestability and market structure in European banking

sectors on the eve of EMU. Journal of Banking and Finance 24, 10451066.Gardener, E.P.M., Molyneux, P., 1990. Changes in Western European Banking. Allen Unwin, London.

Gual, J., Neven, D., 1993. Banking. Social EuropeMarket Services and European Integration.

European Economy, Report and Study no. 3.

Lloyd-Williams, D.M., Molyneux, P., Thornton, J., 1991. Competition and contestability in the Japanese

commercial banking market. Institute of European Finance, Research Papers in Banking and Finance,

no. 16, Bangor.

Molyneux, P., Lloyd-Williams, D.M., Thornton, J., 1994. Competitive conditions in European banking.

Journal of Banking and Finance 18, 445459.

Molyneux, P., Altunbas, Y., Gardener, E.P.M., 1996. Efficiency in European Banking. Wiley, New York.

Nathan, A., Neave, E.H., 1989. Competition and contestability in Canadas financial system: Empirical

results. Canadian Journal of Economics 22, 576594.

Panzar, J.C., Rosse, J.N., 1987. Testing for monopoly equilibrium. Journal of Industrial Economics 35,

443456.

Resti, A., 1997. Evaluating the cost-efficiency of the Italian banking system: What can be learnt from the

joint application of parametric and non-parametric techniques. Journal of Banking and Finance 21,

221250.

Rime, B., 1999. Mesure de degree de concurrence dans le systeeme bancaire Suisse aa laide du modeele de

Panzar et Rosse. Revue Suisse dEconomie Politique et de Statistique 135, 2140.

Rosse, J.N., Panzar, J.C., 1977. Chamberlin vs Robinson: An empirical study for monopoly rents. Bell

Laboratories Economic Discussion Paper.

Shaffer, S., 1982. A non-structural test for competition in financial markets. In: Proceedings of a

Conference on Bank Structure and Competition. Federal Reserve Bank of Chicago, pp. 225243.

Shaffer, S., DiSalvo, J., 1994. Conduct in banking duopoly. Journal of Banking and Finance 18, 1063

1082.

Toolsema, L.A., 2002. Competition in the Dutch consumer credit market. Journal of Banking and Finance

26 (this issue).

Vesala, J., 1995. Testing for competition in banking: Behavioural evidence from Finland. Bank of Finland

Studies, E:1.


competition concentration and their relationship an empirical analysis of the banking industry...

Documents